J Happiness Stud (2014) 15:937–956

DOI 10.1007/s10902-013-9457-4

RESEARCH PAPER

Economic Consequences of Mispredicting Utility

Bruno S. Frey • Alois Stutzer

Published online: 2 August 2013

 Springer Science+Business Media Dordrecht 2013

Abstract In a simple conceptual framework, we organize a multitude of phenomena

related to the (mis)prediction of utility. Consequences in terms of distorted choices and
lower well-being emerge if people have to trade-off between alternatives that are char-
acterized by attributes satisfying extrinsic desires and alternatives serving intrinsic needs.
Thereby the neglect of asymmetries in adaptation is proposed as an important driver. The
theoretical analysis is consistent with econometric evidence on commuting choice using
data on subjective well-being. People show substantial adaptation to a higher labor income
but not to commuting. This may account for the finding that people are not compensated
for the burden of commuting.

Keywords Adaptation  Extrinsic/intrinsic attributes  Individual decision-making 

Misprediction  Subjective well-being  Time allocation

JEL Classification A12  D11  D12  D84  I31  J22

1 The Issue

Consider a job offer promising an increase in income of $30,000 a year, but which is
farther away from where you live. Due to a longer commute, you will have less time

B. S. Frey
Department of Economics, University of Zurich, Hottingerstrasse 10, 8032 Zurich, Switzerland
e-mail: [email protected]

B. S. Frey  A. Stutzer
CREMA—Center for Research in Economics, Management and the Arts, Basel, Switzerland

A. Stutzer (&)
Faculty of Business and Economics, University of Basel, Peter Merian-Weg 6, 4002 Basel, Switzerland
e-mail: [email protected]

123
938 B. S. Frey, A. Stutzer

available for your family and friends, hobbies, gardening and dancing. How will you
decide?
We argue that individuals systematically mispredict utility1 in such choice situations.
They underestimate the utility relating to aspects of consumption satisfying intrinsic needs
(time spent with family and friends or on hobbies). In contrast, the characteristics relating
to consumption satisfying extrinsic desires (income and status) are overvalued. As a
consequence, individuals tend to under-consume goods and activities with strong intrinsic
attributes, compared to those with strong extrinsic attributes. According to their own
subjective evaluation, individuals make distorted decisions when they choose between
different options and obtain a lower utility level than they otherwise would.
The argument is based on the notion that individuals find it difficult to make compar-
isons between attributes, whose salience shifts over time. Specifically, people fail to make
an accurate prediction of adaptation. Moreover, adaptation and the degree of its prediction
systematically differ across goods and activities. As a differentiation criterion, we propose
the nature of needs—either being intrinsic or extrinsic—that choices are satisfying. The
emphasis is thus on the intrinsic and extrinsic attributes of choice options. It is hypothe-
sized that people adapt less to intrinsically rewarding activities and goods than to extrinsic
satisfiers.2
We derive two propositions from our arguments and empirically test them in an
application to commuting decisions. First, we study whether people commute too much in
the sense that they are not fully compensated for the burden of commuting. Second, we
analyze whether there is differential adaptation to a higher labor income and an increase in
commuting time.3 Our findings, based on panel data on reported life satisfaction for
Germany, show substantial adaptation to a higher labor income, but sensitization, albeit
small, to commuting. This result can account for the paradoxical observation that people
are not fully compensated for the burden of commuting.
This paper intends to make three contributions. First, a simple conceptual framework is
proposed to organize a multitude of phenomena related to the prediction of utility in
research in economics and psychology. The framework should make it easier to integrate
these phenomena into economics. Second, based on the categorization of goods and
activities as predominantly characterized by extrinsic or intrinsic attributes, testable pre-
dictions are derived about the economic consequences in terms of distorted choices. The
categorization of choice attributes is considered the main novelty of the paper. The

1
The misprediction of utility in general has recently been introduced in the series on anomalies in the
Journal of Economic Perspectives (Kahneman and Thaler 2006). Specific suboptimal consumption choices
are reviewed in Hsee et al. (2012). It is argued, for example, that the mode under which available choice
options are evaluated is a source of error when people predict the utility they may accrue from the options.
2
Our proposition is related to a long tradition in economics arguing that individuals tend to focus too much
on material goods and disregard goods providing non-material benefits (see Lebergott 1993; Lane 1991;
Frank 1999). Most importantly, Scitovsky (1976) claimed that ‘‘comfort goods’’ are over-consumed com-
pared to goods providing ‘‘stimulation’’ (for a discussion of Scitovsky’s contribution to the understanding of
well-being, see Pugno 2013). Comfort goods are described as defensive activities, protecting against neg-
ative affect. They have a strong extrinsic component, consisting of the consumer goods achieved through
rapid productivity growth. In contrast, stimulation comes from creative activities providing novelty, sur-
prise, variety and complexity. These aspects accentuate the renewal of pleasurable experiences, as it is also
emphasized for intrinsic attributes.
3
Taking commuters’ reported satisfaction with life as a proxy measure for individually experienced utility
is an approach that follows a substantial recent literature on reported subjective well-being, satisfaction and
happiness in economics (e.g. Frey and Stutzer 2002; Layard 2005; Di Tella and MacCulloch 2006; Stutzer
and Frey 2010), as well as in psychology (e.g. Kahneman et al. 1999; Diener et al. 1999).

123
Economic Consequences of Mispredicting Utility 939

differentiation drives the main implications of our analysis. Third, an empirical strategy is
developed, based on data on reported subjective well-being to test for utility misprediction.
This strategy, which focuses on evaluative measures of individual well-being, is set up as a
general strategy that can be applied beyond the specific analysis in this paper. Other
applications might involve decisions about working hours or the timing of when to have
children.
Section 2 discusses individual decision-making, where the salience of the characteris-
tics of goods and activities differs from the moment when people make a decision to the
period of consumption. Section 3 gives reasons why people mispredict utility and under-
value the intrinsic attributes of choice options when compared to the extrinsic attributes.
We draw on both psychological and institutional insights and refer to related phenomena.
The following Sect. 4 raises the question why there is little or no learning in utility
misprediction. Section 5 provides an econometric analysis for a specific, but important
example; the income/commuting trade-off discussed at the outset. Section 6 draws con-
cluding remarks.

2 Individual Decision-Making When the Salience of Attributes Changes

Standard economic theory assumes that individuals are able to compare the future utilities
provided by the goods and activities consumed, and that they maximize their own utility in
a rational consumption decision. In certain cases, it has proved useful to distinguish
between the various characteristics of goods and activities (Lancaster 1966; Becker 1965)
or the different attributes of options (e.g. Keeney and Raiffa 1976). However, this dif-
ferentiation is not taken to affect the evaluation of future utility. The utility of a chosen
combination is simply the sum of the weighted value of each characteristic.
The standard economic model of consumer decisions is appropriate for most goods and
activities and for most situations.4 It is also appropriate when individuals make random
prediction errors, or when the extent of misprediction is the same for all goods and all
activities.
This paper diverges from the assumptions of the revealed preference approach under-
lying standard economics in two significant ways:
First, we allow for the fact that people mispredict utility; i.e. ex ante predicted utility
from different choice options (that is supposed to guide people’s behavior) does not match
with actual ex post experienced utility.5 In order to generate a tractable framework, we
simply represent this misprediction as a change in the salience (or the weight) given to
different attributes of goods and activities in people’s evaluations (ex ante and ex post).6
Different sources of utility misprediction are discussed in the next section.
Second, we argue that there are systematic differences in misprediction and focus on the
attributes of choice options. Only with this additional presumption is a directed hypothesis

4
A characterization of the economic approach to human behavior is provided, for example, in Frey and
Stutzer (2001).
5
This terminology follows the work of Kahneman et al. (1997) in which utility is interpreted as a hedonic
experience. Both utility measures—predicted and experienced utility—diverge from traditional decision
utility that is derived from individual behavior.
6
A theory of the role of salience in choice under risk is proposed in Bordalo et al. (2012). In their model,
those attributes of choice options are salient that are dissimilar across options.

123
940 B. S. Frey, A. Stutzer

about distorted behavior possible. In this second conceptual step, we differentiate between
two types of attributes characterizing various options.7
• Attributes of the first type relate to ‘intrinsic needs’. A comprehensive view of the main
aspects of intrinsic needs is provided in the self-determination theory of Deci and Ryan
(2000). First, there is a need for relatedness, referring to the desire to feel connected to
others through love and affection (having family and friends and being in a social
setting). Second, a need for competence refers to the propensity to control the
environment and experience oneself as capable and effective. Third, a desire for
autonomy involves the experience of being in charge of one’s actions or being causal.8
Intrinsic need attributes are also characterized as providing ‘‘flow experience’’
(Csikszentmihalyi 1990), i.e. when one is completely immersed in an activity, often a
hobby.
• The second type of attributes relates to ‘extrinsic desires’. Extrinsic attributes serve
people’s goals for material possessions, fame, status or prestige. Income thus becomes
a crucial aspect of options in the choice set. A high income allows for a high standard
of living in material terms.9
Each option, activity and even good is multidimensional; in general, a particular choice
alternative has both intrinsic and extrinsic need attributes or, in short, intrinsic and extrinsic
attributes. But some goods and activities are more intrinsic by nature. Examples might be
time spent with friends,10 voluntary work, playing a musical instrument, attending a
concert, repairing a broken household tool, solving a problem at work, or providing
professional services to a customer’s satisfaction. There are other goods and activities that
are more extrinsic in nature. These include, for example, consumer articles that serve a
purpose beyond basic material needs (like designer clothes or consumer electronics) or
involve a strong status component (like cars of different brands). Other examples are
conspicuous job benefits (like laptops) or trips to places that one has to have visited. In this
analysis, we neglect the satisfaction of physiological needs and concentrate on the
gratification of needs through time and income that are available for discretionary use.
Our main proposition is that, when making a decision, the extrinsic attributes are
relatively more salient than the intrinsic attributes of different options. When it comes to
decision-making, individuals therefore tend to undervalue the future utility of intrinsic
attributes compared to extrinsic attributes. This distortion leads to a systematic discrepancy
between predicted utility and experienced utility.11

7
We borrow these categories from a large literature in humanistic or value psychology (e.g. Maslow 1968;
Rogers 1961).
8
The underlying theories are manifold, and include the urge to master one’s environment for its own sake
(White 1959) to be an origin (DeCharms 1968), to resist loss of control (Brehm 1966) and the reflection of
perceived control in more effective behavior and higher positive affects (Bandura 1977; Seligman 1992).
9
Alternative classifications of attributes and characteristics might well be possible and productive. For
example, in a recent application, pro-environmental attributes of choice options are linked to utility mis-
prediction (Welsch and Kühling 2011).
10
When people spend time with friends because they are famous or important, the extrinsic dimension
becomes more prevalent.
11
A complementary discrepancy is the one between the happy life filled with positive affect and the
meaningful life (Baumeister et al. 2013).

123
Economic Consequences of Mispredicting Utility 941

3 Potential Sources for Mispredicting Utility

There is a range of theoretical concepts and empirical phenomena in economics and

psychology that can be reconciled with our conceptual framework. We consider the pro-
posed mechanisms as potential sources for mispredicting utility.

3.1 Underestimating (Asymmetric) Adaptation

There is convincing empirical evidence that individuals are not good at foreseeing how
much utility they will derive from future consumption (e.g. Loewenstein and Adler 1995).
For surveys, see Wilson and Gilbert (2003) or Gilbert and Wilson (2009). Research on
affective forecasting shows, in particular, that people tend to overestimate their reactions to
specific events because they are embedded within other daily life events that they are not
currently aware of: for instance, seeing their favorite soccer team win is experienced
simultaneously with other events (like quarreling with one’s partner or unexpectedly being
given additional professional work). Another example of errors in predicting emotions is
that people underestimate their ability to successfully cope with negative events.12 The
general notion is that people usually have biased expectations about the intensity and
duration of emotions, in the sense that the emotional impact is often lower than predicted
because people adapt more than they foresee. Loewenstein et al. (2003) neatly introduce
this notion in a theoretical model of intertemporal decision-making.
We argue that adaptation is more likely to be underestimated for extrinsic aspects than
for intrinsic aspects. People adapt less to goods and activities with strong intrinsic com-
ponents because the (positive) experience tends to be renewed with every new act of
consumption. We claim that getting together with a good friend is always rewarding, and
one does not get used to it in the sense of valuing this experience less and less. Rather, the
opposite is true. Each interaction with the friend provides fresh pleasure and enjoyment.
Similarly, many scholars have a flow experience when they immerse themselves in writing
a paper or book they always wanted to write. The corresponding utility does not wear off.
Thus, many senior scholars, who have written numerous papers and books in the past,
experience the same flow as when they were young.
These aspects of adaptation fit well the AREA model of Wilson and Gilbert (2008). In
their model, the process of affective adaption is determined by people’s attempt to
understand events that attract their attention. Thereby, people attend and emotionally react
to unexplained events that are relevant for their self. If they are successful and understand
the events, they give them less attention and the affective reactions to them get weaker
(p. 371). Key variables that impede explanation are an event’s novelty, unexpectedness,
variability, uncertainty and explanatory incoherence. These aspects often exactly charac-
terize attributes of choice options serving intrinsic needs as exemplified above. Moreover,
there is generally more attention (and thus less adaptation) attributed to events that matter
for people’s identity or self fitting well the distinction from self-determination theory.

12
Young academics might be particularly worried about life after a negative tenure decision. Gilbert et al.
(1998) asked assistant professors how happy they thought they would be after a positive and a negative
tenure decision. The answers were compared with the reported subjective well-being of academics affected
by a tenure decision made five or less years previously. Although assistants predicted they would be less
happy in the first five years after being turned down, there was no statistically significant difference between
those who had and had not gotten tenure. Similarly, assistants also overestimated the positive impact of
receiving tenure on their subjective well-being.

123
942 B. S. Frey, A. Stutzer

The claim of differential adaptation to intrinsic and extrinsic attributes of goods and
activities is also consistent with much of the recent empirical evidence (for surveys, see
Frederick and Loewenstein 1999; Luhman et al. 2012). It has been found that individuals do
not adapt their utility evaluation in the case of undesirable experiences that inhibit intrinsic
need satisfaction. In particular, severe health problems, like chronic illness, or an illness that
gets progressively worse, reduce autonomy and lead to lasting reductions in reported sub-
jective well-being (e.g. Easterlin 2003; Oswald and Powdthavee 2008). People also do not
fully adapt to positive intrinsic need fulfillment. This is, the case, for example, for marriage
supporting the desire to be connected to others through love and affection (Clark et al. 2008;
Stutzer and Frey 2006). Furthermore, there is limited adaptation to negative shocks to rela-
tional goods, in particular to the death of close relatives (e.g. Frijter et al. 2011). Having a job
is related to many aspects that provide flow experiences and satisfy intrinsic needs, like being
in the company of workmates, applying expertise and experiencing autonomy. Accordingly,
being unemployed is repeatedly found to have high negative non-pecuniary effects on peo-
ple’s subjective well-being, with little habituation (Clark et al. 2001). By way of contrast,
having a job with a high degree of autonomy, as in the case of self-employed people, is related
to high job satisfaction. Thus Benz and Frey (2004) show that self-employed people derive
more utility from their work than people employed by an organization, irrespective of income
earned or hours worked. Moreover, they can explain this difference by people’s evaluation of
the use of initiative at their work place and their satisfaction with the actual work itself.
Intrinsic attributes also characterize volunteer work. It is found that people doing volunteer
work are more satisfied with their life in general, even when the possibility of reverse
causality is accounted for (Meier and Stutzer 2008).
In contrast, there is empirical evidence that individuals experience a considerable extent
of adaptation in the case of goods and activities in which the extrinsic aspects are domi-
nant. In particular, this has been demonstrated for income (van Praag 1993; Easterlin 2001;
Stutzer 2004; Di Tella et al. 2010). When individuals experience a rise in income, their
utility level at first increases but then decreases again. It has been estimated (van Her-
waarden et al. 1977) that around 60 % of the utility increase due to a higher position in the
income distribution disappears over time.
The evidence of little adaptation for goods and activities characterized by intrinsic
aspects, and strong adaptation for those characterized by extrinsic aspects, suggests that
individuals who underestimate adaptation, or even disregard adaptation altogether, tend to
make a bigger mistake when predicting future utility from extrinsic attributes than from
intrinsic attributes.

3.2 Other Potential Sources for Mispredicting Utility

3.2.1 Distorted Memory of Past Experiences

When individuals—in the absence of information on current experience—make decisions

about future consumption, or allocation of time, they have to refer to past experiences. People
reflect on specific moments from the past or access generalizations about likely emotions in a
particular type of situation (for a discussion, see Robinson and Clore 2002). If specific
information is available, it has priority in people’s judgment. Thereby, the more memorable
moments of an experience disproportionately affect retrospective assessments of feelings
(Kahneman 1999). What counts as ‘‘more memorable’’ tends to be the most intense moment
(peak) and the most recent moment (end) of an emotional event. This peak-end rule or
duration neglect has been established in many experimental tests (Kahneman 2003).

123
Economic Consequences of Mispredicting Utility 943

Intrinsic attributes are seen to relate to long-term experiences of moderate but enduring
positive feelings. In order to be open to renewed enjoyment of the type of interactions
mentioned above, as well as to be able to immerse oneself in a flow experience, time is
needed. In contrast, extrinsic attributes are related to short-term experiences, in particular
peak emotions. As a result, we argue that the intrinsic aspects of goods and activities
related to their duration (compared to the extrinsic aspects related to peaks) are underes-
timated when people predict utility based on retrospection.

3.2.2 Rationalization of Decisions

Individuals have a strong urge to justify their decisions, both to themselves and to other
people (for pre-decision justification, see Shafir et al. 1993). Not only does predicted
consumption utility affect, for instance, consumption decisions, but also whether people
think that they are getting a bargain (Thaler 1999). There is a general tendency to resist
affective influences and to take rationalistic attributes into account when making decisions.
Hsee et al. (2003) call this reason-based choice ‘‘lay rationalism’’. In experiments, they
find, for instance, that people focus their decisions on absolute economic payoffs and play
down non-economic concerns. Such asymmetries might apply, for example, when com-
paring jobs with different pay and different working atmospheres, whereby the latter are
de-emphasized. Similarly, in the assessment of housing, the price per m2 of living space is
presumably a rational attribute, while feelings about the new neighbors are not. Other
experiments find that people seem to base their choices on rules and principles, and to
bypass predictions on the experiential consequences of their choices (e.g. Prelec and
Herrnstein 1991). These arguments imply that people do not optimally consider the various
attributes of different options so that predicted utility would be maximized.13
We argue that, for extrinsic and intrinsic attributes, there is a similar inconsistency when
it comes to decision-making. It is much easier to provide rationalistic justifications for
extrinsic rather than intrinsic characteristics. Consider again the job offer providing more
income but less leisure-time. Most people will find it much easier to justify both to
themselves and to others why they should accept the job offer, as the extrinsic monetary
dimension is salient. In contrast, it is quite difficult to justify why the intrinsic charac-
teristics provided by more leisure-time (even when its hedonic utility might be correctly
predicted) are important enough to refuse the large increase in money. As a result, goods
and activities characterized by strong intrinsic attributes tend to carry little weight when it
comes to decision-making, compared to extrinsic components.

3.2.3 Intuitive Theories about the Sources of Utility

People have very diverse intuitive theories about what makes them happy (for a discussion
see Loewenstein and Schkade 1999). These beliefs have a direct influence on people
predicting future utility and can cause them to make mistakes. Moreover, these beliefs play
a role because they shape the reconstruction of past emotions and make them consistent
with current self-conceptions or beliefs (Ross 1989). Thus, intuitive theories interact with
the three previously discussed sources of misprediction.

13
The arguments further complement the research on heuristics or ‘‘rules of thumbs’’. The idea is that
adopting them is leading to better decisions than evaluating expected outcomes on a case-by-case basis. The
crucial challenge is the identification of conditions under which the use of rules and principles leads to
favorable outcomes (for a discussion see Hsee et al. 2012).

123
944 B. S. Frey, A. Stutzer

An important belief refers to acquisition and possession as important goals on the way
to happiness, i.e. materialism (e.g. Tatzel 2002 for a discussion in economics). It is found
that people with material or extrinsic life goals report lower self-esteem and life satis-
faction than people with intrinsic life goals (e.g. Kasser and Ryan 1996; Sirgy 1997). This
correlation is probably partly due to confounding factors, like unobserved personality traits
and reversed causality due to a compensatory reaction of people with low subjective well-
being. However, it might also indicate that people who believe intuitively in extrinsic
attributes are prone to mispredict future utility. In contrast, people with intrinsic life goals
for personal growth, relationships and community spirit apply intuitive theories that
emphasize intrinsic attributes, which in turn lead to fewer mispredictions in future utility.
Our argument thus includes heterogeneity among individuals that leads to additional
testable predictions when combined with previous reasons for misprediction.

3.2.4 Institutional Conditions

The differential effect of misprediction between intrinsic and extrinsic attributes also depends
on the extent to which the market enters into the matter. The monetization of a good or activity
induces individuals to focus more on extrinsic attributes than they otherwise would. This
applies to both work and consumption. It has been argued that introducing pay for perfor-
mance leads employees to regard those performance aspects, which are relevant for the
compensation they receive, as predominant. In contrast, aspects of performance irrelevant to
pay are crowded-out (see Frey 1997 and, for a survey of empirical evidence, Frey and Jegen
2001). In the area of consumption, advertising is often aimed at extrinsic aspects of the goods
to be sold. In comparison, lobbies for intrinsic values tend to be weak and sometimes do not
exist at all. To the extent to which ‘‘commercialization’’ occurs (see e.g. Kuttner 1997; Lane
1991), individuals are induced to mispredict the future utility of goods. They are led to believe
that the extrinsic characteristics will make them happier than is actually the case compared to
the intrinsic characteristics.
The choice of processes and institutions might itself be susceptible to utility mispre-
diction. Procedural utility, i.e. the satisfaction derived from the process itself rather than
from its outcome, refers to innate needs. The utility derived from a particular process
contributes to competence, relatedness and autonomy, and is therefore closely related to
the intrinsic attributes of goods and activities (see the survey by Frey et al. 2004).
According to our propositions, sources of procedural utility are likely to be underestimated
in people’s decision-making. Consistent with this idea, it has been empirically shown
(Tyler et al. 1999) that, when making decisions, individuals tend to prefer institutions
promising favorable outcomes. But they state ex post that they would have preferred an
institution putting more emphasis on (just) procedures.

4 Why Is There Little or No Learning?

Systematically mispredicting future utilities, even if they differed between goods and
activities, would be of little consequence for economics if individuals would learn quickly
in repetitive choice situations.14 If this were the case, mispredicting would be a

14
In contrast, learning is not an option when it comes to once-in-a-lifetime choices. Biased decisions can
then well affect one’s life path. We believe that misprediction of utility matters greatly when it comes to
important life decisions (like career choice), but we have not studied them here.

123
Economic Consequences of Mispredicting Utility 945

disequilibrium phenomenon, not basically affecting the notion of rational decision-makers

maximizing individual utility.
A large literature suggests, however, that learning is a complex process, which does not
necessarily lead to overcoming mispredictions. Kahneman (2011) offers an excellent
account of the difficulties in learning when different mental processes affect behavior.
In the choice situation considered here, where the importance of various attributes differ
between the point of time when people have to make a decision and actual consumption
time, learning is much more difficult. Where decisions on future consumption are con-
cerned, learning must often be based on the recollection of past feelings. They are therefore
subject to the same misperceptions as remembering the utility of past experiences (see
previous section on distorted memory). Learning is particularly hampered when there are
too few episodic memories and people rely to a large extent on their intuitive theories
(Robinson and Clore 2002). In consequence, remembered utility and predicted utility
become similar and relatively independent of the utility actually experienced.
Learning, in contrast, is easier when people can access their feelings directly, i.e. while
still experiencing a particular situation. It might even inspire them to adopt institutional
preconditions to sustain optimal decisions after the event. Most of us are familiar with the
experience of not getting together with friends as often as we would really like when
reflecting on it immediately after the meeting. It is difficult to imagine how enjoyable it
was once we are back in our daily routine and have to find time in our busy schedules. One
of the authors experimented with trying to overcome this particular problem by fixing a
new date whilst still with the friends and aware of the pleasure of being in their company. It
resulted in getting together more frequently and enjoying the meetings to the same extent
as before. On another note, moments of pure bliss and very traumatic experiences can
abruptly change people’s intuitive theories about what constitutes happiness.
In general, however, a more elaborate learning process is required. The individual needs
to step back from his or her actual decision-making activity, where the extrinsic charac-
teristics dominate over the intrinsic characteristics. He or she should attempt to make an
overall evaluation, including undertaking some critical self-examination. As such elaborate
learning is more costly, and is itself subject to errors, individuals are not able to fully
correct their mispredictions within a reasonably short period of time. In many cases, they
are not capable of making any correction, so that the misprediction of future utilities
persists over time.
Limited learning might well co-exist with people’s partial awareness of themselves or
others mispredicting utility. Many people talk, for example, about the difficulties and
mistakes they experience in balancing their working life with their personal life. Yet, on a
case-by-case basis, they still make decisions underestimating intrinsic attributes relative to
extrinsic attributes.
A more fundamental reason for a person’s limited learning capacity might lie in the
functionality of misprediction in the evolutionary process. Rayo and Becker (2007) as well
as Graham and Oswald (2010) and Robson and Samuelson (2011) model how humans’
utility functions formed in order to motivate striving for improvements or more generally
to maximize success in genetic replication. Their model rationalizes that people neglect
adaptation (described as self-inflicted externality). However, in today’s world, this utility
function with an inbuilt misprediction may no longer be helpful in guaranteeing an optimal
mix between experienced utility and motivation for success in society.

123
946 B. S. Frey, A. Stutzer

5 Empirical Application to Commuting Choice

When people choose between different jobs and different possible places of residence, they
face the difficult decision of how far they are prepared to commute.15 It involves chal-
lenging trade-offs, often along the line that housing further away would be more exclusive
for the same price, or a job further away would come with a higher salary. Based on the
hypothesis that, when making decisions, people systematically underestimate future utility
from the intrinsic attributes of goods and activities compared to extrinsic attributes, it is
possible to predict individuals’ choices concerning different jobs and different places of
residence. He or she can be expected to underestimate an important aspect, namely
commuting time. Time spent commuting is no longer available for spending with friends
and family members or indulging in a favorite hobby, i.e. intrinsically rewarding activities.
People mispredict utility when they choose a job/housing situation and end up spending too
much time commuting. Job and housing choices are often costly to correct, so that people
are trapped afterwards in an unfavorable commuting situation with a burden that is not
compensated.16
In order to empirically test the general hypothesis of ‘‘too much commuting’’, first a
benchmark of optimal commuting is necessary. For this aspect, we draw on and extend our
earlier research (Stutzer and Frey 2008). Second, to substantiate the claim that commuting
time is underestimated, evidence for a specific asymmetry in people’s evaluations is
necessary. In the following, we propose respective empirical tests, introduce the panel data
for Germany and present the results.

5.1 Constructing an Empirical Test for Full Compensation

The prediction in neoclassical economics of individuals making perfectly rational deci-

sions is taken as a reference standard for optimal commuting. In standard economics,
people are assumed to have little difficulty with optimization when goods and activities
have multiple characteristics. Commuting is like any other characteristic. Monetary costs
involved and physical or mental stress17 enter negatively into the evaluation, and have to

15
In an empirical analysis, Simonsohn (2006) provides related evidence for the inherent complexity of
commuting choice. He argues that commuting behavior can be better understood in a framework of con-
structed preferences. People come up with some reference level of commuting time or commuting radius
that they are only prepared to give up after experiencing negative effects on their well-being. In a chal-
lenging study on people moving from one US city to another, Simonsohn finds that people coming from a
city where the average commuting time of the population is high (or low) also choose to commute more (or
less) than average at their new place of residence (keeping individuals’ own past commuting experience
constant). In the latter model, people can end up either commuting too much or too little.
16
In a related study, Comerford (2011) analyzes the choice of transport mode for commuting in a
framework that takes into account focalism. Focalism is a cognitive bias that leads individuals to overweigh
the contribution of certain attributes of choice options to experienced utility.
17
It is well documented that commuting is both physically and mentally stressful (e.g. Novaco et al. 1990).
The strain of commuting is associated with raised blood pressure, musculoskeletal disorders, lowered
frustration tolerance, increased anxiety and hostility, being in a negative mood when going to work in the
morning and coming home in the evening, increased lateness, absenteeism and turnover at work, as well as
adverse effects on cognitive performance (Koslowsky et al. 1995). In a recent panel study for the UK,
Roberts et al. (2011) find commuting to be related to lower psychological health in particular for women.
This effect holds when statistically controlling for health status, net household income, housing quality,
working hours as well as job satisfaction. As the controls include (1) variables that potentially capture
compensation in the housing and labor market and (2) a measure of domain satisfaction (i.e. job satisfac-
tion), the results cannot easily be interpreted within the framework of the current study.

123
Economic Consequences of Mispredicting Utility 947

be offset by other characteristics, like income or lower housing costs. Only then is it likely
that a job offer involving a longer commuting time may be chosen. This basic idea of
compensation is the driving force behind the notion of a spatial equilibrium in urban
location theory (e.g. Alonso 1964; Moses 1962), as well as in public economic theory,
based on Tiebout’s (1956) model of fiscal competition between jurisdictions. Accordingly,
commuting is determined by an equilibrium state of the housing and labor market, in which
people’s utility is equalized over all actual combinations of alternatives in these two
markets.18
Utility U is thus equal to U  for realized combinations of income yi, time spent com-
muting Di and rent ri across individuals indexed by i
Ui ¼ uðyi ; Di ; ri Þ ¼ U 8i: ð1Þ
Totally differentiating this equilibrium condition leads to
ou ou ou
dU ¼ dy þ dD þ dr ¼ 0: ð2Þ
oy oD or
For variation in commuting time D, this implies that
dU ou dy ou ou dr
¼ þ þ ¼ 0: ð3Þ
dD oy dD oD or dD
The left hand side of Eq. (3) states that the overall change in utility due to a change in the
disamenity cost of commuting time is zero. A decomposition of the total change is pro-
vided on the right hand side of Eq. (3). There are three effects from an increase in
commuting time: there is a marginal gain in utility due to a higher level of consumption
that is reached because jobs that require longer commutes offer a higher income. More-
over, longer commuting time reduces house prices/rents for housing and thus leaves
additional money for consumption. Besides these two positive effects, there is a marginal
decrease in utility due to the burden of spending more time commuting. Given that
incomes and rents for housing exclusively reflect compensation for commuting conditions,
the three effects add up to zero.
The prediction in Eq. (3) can be directly tested, provided utility is observable. Stutzer
and Frey (2008) set up such a test, taking commuters’ reported satisfaction with life as a
proxy measure for individually experienced utility. The idea for the empirical test is
captured in the following regression equation
ui ¼ a þ bDi þ ei : ð4Þ
The coefficient b measures the total change in utility due to a change in commuting time.
Under the null hypothesis b = 0, commuting time is entirely compensated either by higher
salaries or by lower rents for housing. This is the prediction of standard economic theory,
assuming full compensation of the cost of commuting through higher income and lower
housing costs. The alternative hypothesis b \ 0 states that commuting time is not fully
compensated on the labor and housing market. b \ 0 is predicted when commuters

18
The strong notion of equilibrium has only been partially tested so far. It has not been studied whether
there are systematic rents: rather, derived hypotheses within the equilibrium framework have been analyzed.
There is considerable evidence for capitalization of transportation infrastructure in the price of land and for
compensating wage differentials due to commuting distance. However, these findings do not require an
equilibrium situation, and can also be explained by the law of marginal substitution (e.g. Timothy and
Wheaton 2001; van Ommeren 2000).

123
948 B. S. Frey, A. Stutzer

systematically underestimate the costs of commuting and accept jobs or choose housing
that do not fully compensate them.
The actual regression estimated below with panel data takes individual heterogeneity
into account. In particular, a large set of covariates X of reported life satisfaction, as well as
individual and time specific effects ki and st are controlled for:
uit ¼ a þ bDit þ uXit þ ki þ st þ eit : ð5Þ
No job or housing related variables (most importantly labor income) are included. This is
crucial, because income is one of the variables through which people are expected to be
compensated for their daily journey to and from work. Equation (5) only makes a clear
prediction of b = 0 if all sources of compensation remain uncontrolled.19

6 Construction of the Empirical Test for Differential Adaptation

A crucial aspect of whether trade-offs lead to systematically distorted decisions because of

utility misprediction rests on some sort of asymmetry in the degree of misprediction across
choice options. We emphasize the differential adaptation to goods and activities that satisfy
intrinsic and extrinsic needs. In the case of commuting, commuting time refers to intrinsic
needs while labor income is one possible variable that refers to extrinsic needs.
Adaptation to repeated stimuli can be studied using multiple reports of a person’s
subjective well-being. As we want to understand misprediction involved in people’s more
or less deliberate choices, we study adaptation to variation in circumstances that are
brought about by the normal course of life rather than by some external shocks. Moreover,
the set of covariates of reported life satisfaction is extended to include working hours and
labor income in order to separate adaptation to the major characteristics of someone’s work
and commuting situation.20 For the empirical analysis, Eq. (5) is extended to include
current and lagged variables21 of the major choice variables, i.e. commuting time, labor
income y and working hours h:
uit ¼ a þ b1 Dit þ b2 Dit1 þ b3 Dit2 þ b4 Dit3
þ c1 lnyit þ c2 lnyit1 þ c3 lnyit2 þ c4 lnyit3 ð6Þ
þ d1 hit þ d2 hit1 þ d3 hit2 þ d4 hit3 þ /Xit þ ki þ st þ eit

Based on the estimated coefficients b1 to b4 and c1 to c4, the degree of dissipation in the
impact of a change in commuting time and labor income on subjective well-being can be
simply calculated by dividing the sum of the coefficients for the lagged variables by the
initial effect, i.e. (b2 ? b3 ? b4)/b1 and (c2 ? c3 ? c4)/c1. To test for full adaptation, the
P P
hypotheses 4l¼1 bl ¼ 0 and 4l¼1 cl ¼ 0 are set up.

19
If income is kept constant, commuting time is expected to enter negatively into the equation.
20
Due to lack of data, we cannot study adaptation to luxury housing.
21
For the lag structure, we draw on the findings reported in Di Tella et al. (2010) on adaptation to
household income. The adaptation process fades out after three lags. Moreover, the results with three lags
are very similar to the results with four lags but allow us to include more observations.

123
Economic Consequences of Mispredicting Utility 949

7 Data and Sample Selection

The two proposed empirical tests are performed with data from the German Socio-Eco-
nomic Panel Study (GSOEP). The GSOEP is one of the most valuable data sets for
studying individual well-being over time. It was started in 1984 as a longitudinal survey of
private households and persons in the Federal Republic of Germany and was extended to
residents in the former German Democratic Republic in 1990. Reported subjective well-
being is based on the question ‘‘How satisfied are you with your life, all things consid-
ered?’’ Responses range on a scale from 0 ‘‘completely dissatisfied’’ to 10 ‘‘completely
satisfied’’. In addition, people in the survey are asked a wide range of questions with regard
to their socio-economic status and their demographic characteristics.
Information on individual commuting time was collected in eight waves between 1985
and 2003: 1985, 1990 and 1993 for the old German Laender, 1992 for the New German
Laender and 1993, 1995, 1998 and 2003 for all the Laender. People were asked ‘‘How long
does it normally take you to go all the way from your home to your place of work using the
most direct route (one way only)?’’ In order to study adaptation based on the formulated
strategy, information on commuting time is required annually. As there is detailed infor-
mation in the GSOEP about changes in people’s jobs or places of residence, values for
commuting time in intermediate years can easily be imputed. We follow a simple proce-
dure. First, we check whether a person either changes his job and/or his place of residence
between the waves for which commuting time is available. For those identified as movers,
imputation of commuting time is possible using information about past and future com-
muting time. As long as they stay in the same job and place of residence, commuting time
is carried forward from the last reported year to the following year(s) where the infor-
mation is missing. Accordingly, for years in the past respondents stayed in the same job
and residence, commuting time is imputed backwards. If someone only moves once
between years with reported information on commuting time, commuting time can be
imputed throughout. For those identified as non-movers, commuting time between years
with reported information is linearly interpolated.
The sample selection is based on the following criteria: First, we restrict the sample to
those who either commute on a regular basis to the same workplace, or work at home, and
report being either employed or self-employed. Second, we focus on people who are most
likely to be equally involved in the household decision on where to live and work, i.e. the
heads of households and their spouses. Third, as underemployment is a serious restriction
to employees in Germany (Statistisches Bundesamt Deutschland 2006), we only include
full time workers in the sample. On average, people in the selected sample commute
22 min one way, with a standard deviation of 18 min. Thereby commuting time is set to
zero for those people who work at home. Commuting time at the first quartile is 15 min, at
the median 20 min, at the third quartile 30 min and at the 90th percentile 45 min.

8 Econometric Estimations for Incomplete Compensation

The first empirical test extends previous work on commuting and life satisfaction by
Stutzer and Frey (2008). The results of the earlier study are summarized and replicated for
the extended data set with imputed data on commuting time. Moreover, they are integrated
in the context of utility misprediction.
In Table 1, Eq. (5), which captures the effect of commuting time on life satisfaction, is
estimated in a least squares regression (column A), taking a large number of individual

123
950 B. S. Frey, A. Stutzer

Table 1 Commuting and life satisfaction: compensation and adaptation

(A) (B) (C)

Commuting time t -0.00190*** -0.00203*** -0.00200

(0.00056) (0.00056) (0.00169)
t-1 -0.00049
(0.00199)
t-2 -0.00010
(0.00184)
t-3 0.00010
(0.00131)
Ln real labor income t Not incl. 0.18283*** 0.15694***
(0.01512) (0.02990)
t-1 -0.02555
(0.02982)
t-2 -0.02317
(0.02911)
t-3 -0.06059**
(0.02890)
Working hours Not incl. In year t incl. In years t - 3 to t incl.
Ind. characteristicsa Incl. Incl. Incl.
Individual fixed effects Incl. Incl. Incl.
Year fixed effects Incl. Incl. Incl.
No. of observations 85,332 85,332 37,368
Data source GSOEP
Full time employed or self-employed people in Germany, 1984–2005. Dependent variable: satisfaction with
life
Partial correlations are from least square estimations. Standard errors are in parentheses
Statistical significance: * 0.1 [ p [ 0.05, ** 0.05 [ p [ 0.01 and *** p \ 0.01
a
Individual control variables include age square, nine variables for marital status, three variables for the
number of children in the household, the square root of the number of household members and indicators for
self-employment, residence in the New German Laender, foreigners with EU nationality, other foreigners
and first interview

characteristics into account, as well as year and individual specific effects.22 The latter
exclude spurious correlations due to time-invariant unobserved characteristics of people
that are systematically correlated with people’s commuting time and reported subjective
well-being.
The partial correlation reveals that commuting time has a negative effect on life sat-
isfaction. People who spend 1 h rather than 0 min commuting (one way) report, on
average, a -0.115 points (t = -3.40) lower level of subjective well-being. For one
standard deviation (i.e. 18 min), the effect is -0.034 (t = -3.40). The size of the com-
muting effect for one standard deviation is one-eighth of the effect of finding a partner for
those being single. Compared to the effect of becoming unemployed (= -0.671) (see

22
Here, only ordinary least squares estimations are reported. Thus, it is implicitly assumed that the answers
can be cardinally interpreted. While the ranking information in reported subjective well-being would require
ordered probit or logit regressions, comparative analyses for GSOEP have shown that it makes virtually no
difference whether responses are treated ordinally or cardinally (Ferrer-i-Carbonel and Frijters 2004). The
11 categories of the dependent variable seem to mitigate potential problems from assuming continuity.

123
Economic Consequences of Mispredicting Utility 951

Stutzer and Frey 2004, Table 4), an increase in commuting time by one standard deviation
(1 h) is about one twentieth (sixth) as bad for life satisfaction.
This result is at odds with the benchmark prediction of standard location theory and the
implicit assumption in many economic models that people are compensated, on average,
for commuting. Rather, it is consistent with the view that people who mispredict utility and
opt for too long a commuting time are trapped in a situation in which they are not
compensated for the burden of commuting.
A discussion of the results for the socio-demographic and socio-economic covariates of
life satisfaction in Germany can be found in Stutzer and Frey (2004) and Frijters et al.
(2004).23
However, there might be alternative explanations for commuters not being fully com-
pensated.24 First, although commuting might be a burden for those involved, other
members of the family might benefit so that, overall, the households’ well-being is
equalized. But spouses are not reporting higher life satisfaction when their partners spend
more time commuting. Moreover, the negative correlation is also found for single
households where intra-household bargaining and altruism play much less of a role.
Second, there are search models in urban and regional economics that build in transaction
costs (e.g. Weinberg et al. 1981; van Ommeren et al. 1997). They predict lower utility for
those in a disadvantaged situation with long commuting times (e.g. van Ommeren 2000)
and reckon that transaction costs prevent people from adjusting to economic shocks. In
particular, transaction costs might hinder people who experience a longer or more dis-
turbing commuting time ex post than expected ex ante from re-optimizing. Therefore,
people might be locked into a disadvantaged commuting situation. It is very difficult to
reject an explanation based on transaction costs, especially as transaction costs can easily
systematically interact with our explanation based on utility misprediction. However, an
albeit smaller negative effect of commuting time on life satisfaction is also estimated for
people who either change their job and/or their place of residence and so have the
opportunity of re-optimizing their commuting situation. For them, there might also well be
an explanation in terms of economic costs not yet found and so not yet incorporated into
the analysis. This cost factor would be interesting to know, because it potentially relates to
a sizeable loss in well-being and should be explicitly modeled in urban and public eco-
nomics. Until an adequate rational choice explanation has been provided, we propose the
general result to be consistent with people mispredicting utility.

8.1 Econometric Estimations for Differential Adaptation

The results for the second empirical test based on Eq. (6) are also reported in Table 1. The
impact of the current commuting and working situation on life satisfaction is captured in

23
Here, similar partial correlations for the covariates are estimated. The full equation in panel (A) with
individual fixed effects reads as follows: u = -0.0019*** (SE = 0.0006) * commuting time -0.053 * 10-3
(0.078 * 10-3) * age2 ?0.256*** (0.045) * single with partner ?0.292*** (0.049) * married ?0.340***
(0.100) * separated with partner -0.389*** (0.065) * separated no partner ?0.451*** (0.064) * divorced with
partner -0.024 (0.060) * divorced no partner ?0.623*** (0.165) * widowed with partner -0.093 (0.090) *
widowed no partner -0.446*** (0.131) * spouse living abroad ?0.074*** (0.019) * one child in household
?0.098*** (0.026) * two children in household ?0.157*** (0.041) * three or more children in household
-0.252*** (0.042) * no. of persons in the household1/2 -0.064*** (0.024) * self-employed -0.136 (0.097) *
old German Laender -0.008 (0.110) * EU foreigner ?0.038 (0.079) * other foreigner ?0.187*** (0.025) * first
interview ? year fixed effects ? 7.821*** (0.114).
24
Several of them are empirically addressed in Stutzer and Frey (2008).

123
952 B. S. Frey, A. Stutzer

0.08

Life satisfaction
0.04

(mean set to 0)
0.00
-3 -2 -1 0 1 2 3 4 5

-0.04

-0.08
Year
Adaptation to increase in labor income by 50%
Adaptation to increase in commuting time by 30 minutes

Fig. 1 Adaptation to commuting and labor income. Full time employed or self-employed people in
Germany, 1984–2005. Simulation based on the estimated coefficients in Table 1, panel C. Data source:
GSOEP

column B. In addition to the negative partial correlation for commuting time, there is a
statistically significant positive correlation between real labor income and workers’ well-
being. An increase in labor income by 50 % corresponds to a 0.074 points higher satis-
faction with life.
Current and past levels of commuting time and labor income are taken into account in
column C. The estimated coefficients for the lagged variables allude to a systematically
different pattern. The levels of commuting 1 year back or 2 years back enter negatively
into the estimation and thus hint at sensitization rather than adaptation to commuting. In
contrast, past levels of labor income enter negatively into the equation and indicate
adaptation, i.e. part of the positive impact of a higher income level is offset over time.
Figure 1 shows the differential impact of past experiences graphically. For illustrative
purposes, the path of reported life satisfaction is simulated for an increase in commuting
time by 30 min and a simultaneous income increase of 50 % in t = 0.
The degree of adaptation based on the estimated coefficients for labor income, i.e.
(c2 ? c3 ? c4)/c1, amounts to -0.70 or 70 %.25 The degree of sensitization when expe-
riencing a longer commuting time, (b2 ? b3 ? b4)/b1 is estimated to be around 0.24 or
24 %. The main asymmetry is also reflected when the hypotheses of full adaptation are
P P
tested. 4l¼1 bl ¼ 0 is rejected with Prob(F [ F*) = 0.033. In contrast, 4l¼1 cl ¼ 0 cannot
be rejected [Prob(F [ F*) = 0.333].
In sum, the observed asymmetry in the degree of adaptation to changes in the com-
muting/job situation is consistent with the general claim that adaptation is more pro-
nounced for goods and activities that serve extrinsic needs rather than those that satisfy
intrinsic needs. If people have limited ability to predict adaptation, the asymmetry is
carried forward to the misprediction of utility in commuting choices. The economic
consequence is that people choose work/housing arrangements that involve too much
commuting, for which they are not compensated, and thus experience reduced individual
well-being.

25
The respective estimation for real household post-government income in Di Tella et al. (2010) amounts
to 65.2 %.

123
Economic Consequences of Mispredicting Utility 953

9 Conclusions

This paper argues that individuals systematically mispredict the future utility of the goods
consumed and activities undertaken. Goods and activities characterized by stronger
intrinsic attributes (such as spending time with family and friends and pursuing hobbies)
are undervalued compared to those characterized by stronger extrinsic attributes (such as
most consumer goods). Due to the complexity of having to compare various attributes,
learning is slow and imperfect, so that the distorted decisions are preserved over time. As a
consequence, individuals experience a lower level of utility than if they were not subject to
this systematic bias of misprediction.
The result that individuals are worse off according to their own best interests distin-
guishes us from the more traditional ‘‘consumption critique’’, according to which indi-
viduals are not able to choose what is best for them—but what is ‘‘best’’ is evaluated
according to outside preferences.
The argument synthesizes a multitude of phenomena, mainly from psychology, and
related to utility (mis)prediction in a common framework. Thereby, the asymmetry in the
extent of utility misprediction across goods and activities is emphasized. The proposed
concept should assure the fruitful introduction of utility misprediction into economics in
order to derive predictions about economic consequences in terms of individual behavior
and individual welfare.
In an empirical application, individuals’ commuting decisions are analyzed using data
on subjective well-being. We find that people who spend more time commuting report
lower life satisfaction, i.e. they are not fully compensated for the burden of commuting
either by a higher salary, a better living environment or a lower rent. This is consistent with
people overestimating future utility from the extrinsic attributes of job offers and housing
options, and neglecting the intrinsic attributes, such as the physical burden of commuting
and having less time available to spend with friends and family. In a refined analysis, the
evidence for differences in the degree of adaptation is studied. It is found that full-time
workers adapt, to a large extent, to a higher labor income over a period of 3 years. In
contrast, people adapt much less to commuting, and even seem to become increasingly
sensitive toward the burden of commuting. This latter evidence adds to a consistent
explanation of the commuting phenomenon in terms of utility misprediction. There is an
asymmetry present in adaptation that leads to distorted choices, if adaptation is not taken
into account in the evaluation of alternatives.
Research on utility misprediction in economics is only in the initial stages. Fruitful next
steps could go in many possible directions. While we focus on a specific characterization
of goods and activities, there might be other areas where the degree of utility misprediction
systematically differs. Inspiration might come from the analysis of further trade-offs that
involve choice options that are difficult to evaluate. Further promising studies could be
made. In the work sphere, these studies could be based on the decision to become self-
employed or to do voluntary work. In the private sphere, these studies could focus on TV
viewing, which is the most time-consuming leisure time activity in the Western World, and
involved in many trade-offs in time allocation. Above all, we hope reading this article was
not due to utility misprediction.

Acknowledgments We are grateful for helpful remarks to this and previous versions of the paper from
Roland Bénabou, Matthias Benz, Marina Bianchi, Rafael Di Tella, Ed Diener, Paul Dolan, Richard Eas-
terlin, Bob Frank, Paul Frijters, Klaus Foppa, Ralph Hertwig, Christopher Hsee, Danny Kahneman, Tim
Kasser, David Laibson, Rafael Lalive, Richard Layard, Ed Lazear, Robert MacCulloch, Willi Meyer, Felix

123
954 B. S. Frey, A. Stutzer

Oberholzer, Andrew Oswald, Mirjam Plantinga, Maurizio Pugno, Jason Riis, Wolfgang Stroebe and par-
ticipants at the annual meeting of the American Economic Association, the European Economic Associa-
tion, the German Economic Association and the Swiss Society of Economics and Statistics, as well as at the
International Conferences on Hedonic Adaptation and Prediction in Harvard and on Adaptation and Ref-
erence Values at Brunel Univeristy, the Brookings/Warwick Conference on ‘‘Why Inequality Matters:
Lessons for Policy from the Economics of Happiness’’ in Washington, a public lecture at Princeton
University and the conference ‘‘Economics Meets Psychology’’ in Frankfurt. Data for the German Socio-
Economic Panel has been kindly provided by the German Institute for Economic Research (DIW) in Berlin.

References

Alonso, W. (1964). Location and land use: Toward a general theory of land rent. Cambridge, MA: Harvard
University Press.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavior change. Psychological Review, 84,
191–215.
Baumeister, R. F., Vohs, K. D., Aaker, J. L., & Garbinsky, E. N. (2013). Some key differences between a
happy life and a meaningful life. Journal of Positive Psychology, forthcoming.
Becker, G. S. (1965). A theory in the allocation of time. Economic Journal, 75(299), 493–517.
Benz, M., & Frey, B. S. (2004). Being independent raises happiness at work. Swedish Economic Policy
Review, 11(2), 95–134.
Bordalo, P., Gennaioli, N., & Shleifer, A. (2012). Salience theory of choice under risk. Quarterly Journal of
Economics, 127(3), 1243–1285.
Brehm, J. W. (1966). A theory of psychological reactance. New York: Academic Press.
Clark, A. E., Diener, E., Georgellis, Y., & Lucas, R. (2008). Lags and leads in life satisfaction: A test of the
baseline hypothesis. Economic Journal, 118(529), F222–F243.
Clark, A. E., Georgellis, Y., & Sanfey, P. (2001). Scarring: The psychological impact of past unemploy-
ment. Economica, 68(270), 221–241.
Comerford, D. A. (2011). Attenuating focalism in affective forecast of the commuting experience: Impli-
cation for economic decisions and policy making. Journal of Economic Psychology, 32(5), 691–699.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York: Harper Perennial.
DeCharms, Ri. (1968). Personal causation. New York: Academic Press.
Deci, E. L., & Ryan, R. M. (2000). The ‘‘what’’ and ‘‘why’’ of goal pursuits: Human needs and the self-
determination of behavior. Psychological Inquiry, 11(4), 227–268.
Di Tella, R., Haisken-De New, J., & MacCulloch, R. (2010). Happiness adaptation to income and to status in
an individual panel. Journal of Economic and Behavior Organization, 76(3), 834–852.
Di Tella, R., & MacCulloch, R. (2006). Some uses of happiness data in economics. Journal of Economic
Perspectives, 20(1), 25–46.
Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: Three decades of
progress. Psychological Bulletin, 125(2), 276–303.
Easterlin, R. A. (2001). Income and happiness: Towards a unified theory. Economic Journal, 111(473),
465–484.
Easterlin, R. A. (2003). Building a better theory of well-being. In L. Bruni & P. L. Porta (Eds.), Economics
and happiness: Framing the analysis (pp. 29–64). New York: Oxford University Press.
Ferrer-i-Carbonel, A., & Frijters, P. (2004). The effect of methodology on the determinants of happiness.
Economic Journal, 114, 641–659.
Frank, R. H. (1999). Luxury fever: Why money fails to satisfy in an era of excess. New York: Free Press.
Frederick, S., & Loewenstein, G. (1999). Hedonic adaptation. In D. Kahneman, E. Diener, & N. Schwarz
(Eds.), Well-being: The foundations of hedonic psychology (pp. 302–329). New York: Russell Sage
Foundation.
Frey, B. S. (1997). Not just for the money: An economic theory of personal motivation. Cheltenham, UK and
Brookfield, USA: Edward Elgar.
Frey, B. S., Benz, M., & Stutzer, A. (2004). Introducing procedural utility: Not only what but also how
matters. Journal of Institutional and Theoretical Economics, 160(3), 377–401.
Frey, B. S., & Jegen, R. (2001). Motivation crowding theory: A survey of empirical evidence. Journal of
Economic Surveys, 15(5), 589–611.
Frey, B. S., & Stutzer, A. (2001/2002). Economics and psychology: From imperialistic to inspired eco-
nomics. Revue de Philosophie Economique, 4 5–22.

123
Economic Consequences of Mispredicting Utility 955

Frey, B. S., & Stutzer, A. (2002). What can economists learn from happiness research? Journal of Economic
Literature, 40(2), 402–435.
Frijter, P., Johnston, D. W., & Shields, M. A. (2011). Life satisfaction dynamics with quarterly life event
data. The Scandinavian Journal of Economics, 113(1), 190–211.
Frijters, P., Shields, M. A., & Haisken-DeNew, J. P. (2004). Investigating the patterns and determinants of
life satisfaction in Germany following reunification. Journal of Human Resources, 39(3), 649–674.
Gilbert, D. T., Pinel, E. C., Wilson, T. D., Blumberg, S. J., & Wheatley, T. P. (1998). Immune neglect: A
source of durability bias in affective forecasting. Journal of Personality and Social Psychology, 75(3),
617–638.
Gilbert, D. T., & Wilson, T. D. (2009). Why the brain talks to itself: Sources of error in emotional
prediction. Philosophical Transaction of the Royal Society B, 364, 1335–1341.
Graham, L., & Oswald, A. J. (2010). Hedonic capital, adaption and resilience. Journal of Economic
Behavior and Organization, 76(2), 372–384.
Hsee, C. K., Rottenstreich, Y., & Stutzer, A. (2012). Suboptimal choice and the need for experienced
individual well-being in economic analysis. International Journal of Happiness and Development,
1(1), 63–85.
Hsee, C. K., Zhang, J., Yu, F., & Xi, Y. (2003). Lay rationalism in decision making. Journal of Behavioral
Decision Making, 16, 257–272.
Kahneman, D. (1999). Objective happiness. In D. Kahneman, E. Diener, & N. Schwarz (Eds.), Well-being:
The foundations of hedonic psychology (pp. 3–25). New York: Russell Sage Foundation.
Kahneman, D. (2003). Experienced utility and objective happiness: A moment-based approach. In I. Brocas
& J. D. Carrillo (Eds.), The psychology of economic decisions, volume 1: Rationality and well-being
(pp. 187–208). Oxford: Oxford University Press.
Kahneman, D. (2011). Thinking, fast and slow. New York: Farrar, Strauss, Giroux.
Kahneman, D., Diener, E., & Schwarz, N. (Eds.). (1999). Well-being: The foundation of hedonic psychology.
New York: Russell Sage Foundation.
Kahneman, D., & Thaler, R. H. (2006). Anomalies: Utility maximization and experienced utility. Journal of
Economic Perspectives, 20(1), 221–234.
Kahneman, D., Wakker, P. P., & Sarin, R. (1997). Back to bentham? Explorations of experienced utility.
Quarterly Journal of Economics, 112(2), 375–405.
Kasser, T., & Ryan, R. M. (1996). Further examining the American dream: Differential correlates of
intrinsic and extrinsic goals. Personality and Social Psychology Bulletin, 22(3), 280–287.
Keeney, R. L., & Raiffa, H. (1976). Decisions with multiple objectives. New York: Wiley.
Koslowsky, M., Kluger, A. N., & Reich, M. (1995). Commuting stress: Causes, effects, and methods of
coping. New York: Plenum Press.
Kuttner, R. (1997). Everything for sale: The virtues and limits of markets. New York: Knopf.
Lancaster, K. J. (1966). A new approach to consumer theory. Journal of Political Economy, 74(2), 132–157.
Lane, R. E. (1991). The market experience. Cambridge: Cambridge University Press.
Layard, R. (2005). Happiness: Lessons from a new science. New York: Penguin.
Lebergott, S. (1993). Pursuing happiness: American consumers in the twentieth century. Princeton, NJ:
Princeton University Press.
Loewenstein, G., & Adler, D. (1995). A bias in the prediction of tastes. Economic Journal, 105, 929–937.
Loewenstein, G., O’Donoghue, T., & Rabin, M. (2003). Projection bias in predicting future utility. Quar-
terly Journal of Economics, 118(4), 1209–1248.
Loewenstein, G., & Schkade, D. (1999). Wouldn’t it be nice? Predicting future feelings. In D. Kahneman, E.
Diener, & N. Schwarz (Eds.), Well-being: The foundation of hedonic psychology (pp. 85–105). New
York: Russell Sage Foundation.
Luhman, M., Hofmann, W., Eid, M., & Lucas, R. E. (2012). Subjective well-being and adaptation to life
events: A meta-analysis. Journal of Personality and Social Psychology, 102(3), 592–615.
Maslow, A. (1968). Toward a psychology of being (2nd ed.). New York: Van Nostrand.
Meier, S., & Stutzer, A. (2008). Is volunteering rewarding in itself? Economica, 75(1), 39–59.
Moses, L. N. (1962). Towards a theory of intra-urban wage differentials and their influence on travel
patterns. Papers and Proceedings of the Regional Science Association, 9, 53–63.
Novaco, R. W., Stokols, D., & Milanesi, L. C. (1990). Subjective and objective dimensions of travel
impedance as determinants of commuting stress. American Journal of Community Psychology, 18,
231–257.
Oswald, A., & Powdthavee, N. (2008). Does happiness adapt? A longitudinal study of disability with
implications for economists and judges. Journal of Public Economics, 92(5–6), 1061–1077.
Prelec, D., & Herrnstein, R. J. (1991). Preferences or principles: Alternative guidelines for choice. In
R. J. Zeckhauser (Ed.), Strategy and choice. Cambridge: MIT Press.

123
956 B. S. Frey, A. Stutzer

Pugno, M. (2013). Scitovsky and the income-happiness paradox. Journal of Socio-Economics, 43(2), 1–10.
Rayo, L., & Becker, G. S. (2007). Evolutionary efficiency and happiness. Journal of Political Economy,
115(2), 302–337.
Roberts, J., Hodgson, R., & Dolan, P. (2011). ‘‘It’s driving her mad’’: Gender differences in the effects of
commuting on psychological health. Journal of Health Economics, 30(5), 1064–1076.
Robinson, M. D., & Clore, G. L. (2002). Belief and feeling: Evidence for an accessibility model of
emotional self-report. Psychological Bulletin, 128(6), 934–960.
Robson, A., & Samuelson, L. (2011). The evolution of decision and experienced utilities. Theoretical
Economics, 6(3), 311–339.
Rogers, C. (1961). On becoming a person. Boston: Houghton Mifflin.
Ross, M. (1989). Relation of implicit theories to the construction of personal histories. Psychological
Review, 96, 341–357.
Scitovsky, T. (1976). The joyless economy: An inquiry into human satisfaction and dissatisfaction. Oxford:
Oxford University Press.
Seligman, M. E. P. (1992). Helplessness: On depression, development, and death. New York: Freeman.
Shafir, E., Simonson, I., & Tversky, A. (1993). Reason-based choice. Cognition, 49(2), 11–36.
Simonsohn, U. (2006). New-yorkers commute more everywhere: Contrast effects in the field. Review of
Economics and Statistics, 88(1), 1–9.
Sirgy, M. J. (1997). Materialism and quality of life. Social Indicators Research, 43(3), 227–260.
Statistisches Bundesamt Deutschland (2006). Unterbeschäftigung nimmt zu: Jeder siebte Erwerbstätige
möchte mehr Arbeit. Press Release No. 131, 24 Mar 2006.
Stutzer, A. (2004). The role of income aspirations in individual happiness. Journal of Economic Behavior
and Organization, 54(1), 89–109.
Stutzer, A., & Frey, B. S. (2004). Reported subjective well-being: A challenge for economic theory and
economic policy. Schmollers Jahrbuch, 124(2), 1–41.
Stutzer, A., & Frey, B. S. (2006). Does marriage make people happy, or do happy people get married?
Journal of Socio-Economics, 35(2), 326–347.
Stutzer, A., & Frey, B. S. (2008). Stress that doesn’t pay: The commuting paradox. Scandinavian Journal of
Economics, 110(2), 339–366.
Stutzer, A., & Frey, B. S. (2010). Recent advances in the economics of individual subjective well-being.
Social Research: An International Quarterly, 77(2), 679–714.
Tatzel, M. (2002). ‘‘Money worlds’’ and well-being: An integration of money dispositions, materialism and
price-related behavior. Journal of Economic Psychology, 23, 103–126.
Thaler, R. H. (1999). Mental accounting matters. In D. Kahneman & A. Tversky (Eds.), Choices, values and
frames (pp. 241–268). Cambridge: Cambridge University Press.
Tiebout, C. M. (1956). A pure theory of local expenditure. Journal of Political Economy, 64(5), 416–424.
Timothy, D., & Wheaton, W. C. (2001). Intra-urban wage variation, employment location, and commuting
times. Journal of Urban Economics, 50(2), 338–366.
Tyler, T. R., Huo, Y. J., & Lind, E. A. (1999). The two psychologies of conflict resolution: differing
antecedents of pre-experience choices and post-experience evaluations. Group Processes and Inter-
group Relations, 2(2), 99–118.
van Herwaarden, F., Kapteyn, A., & van Praag, B. M. S. (1977). Twelve thousand individual welfare
functions: A comparison of six samples in Belgium and The Netherlands. European Economic Review,
9(3), 283–300.
van Ommeren, J. (2000). Commuting and relocation of jobs and residences. Aldershot: Ashgate.
van Ommeren, J., Rietveld, P., & Nijkamp, P. (1997). Commuting: In search of jobs and residence. Journal
of Urban Economics, 42, 402–421.
van Praag, B. M. S. (1993). The relativity of the welfare concept. In M. Nussbaum & A. K. Sen (Eds.), The
quality of life (pp. 362–416). Oxford: Clarendon.
Weinberg, D. H., Friedman, J., & Mayo, S. K. (1981). Intraurban residential mobility: The role of trans-
actions costs, market imperfections, and household disequilibrium. Journal of Urban Economics, 9(3),
332–348.
Welsch, H., & Kühling, J. (2011). Are pro-environmental consumption choices utility-maximizing? Evi-
dence from subjective well-being data. Ecological Economics, 72, 75–87.
White, R. W. (1959). Motivation reconsidered: The concept of competence. Psychological Review, 66(5),
297–333.
Wilson, T. D., & Gilbert, D. T. (2003). Affective forecasting. In M. Zanna (Ed.), Advances in experimental
social psychology (Vol. 35, pp. 345–411). New York: Elsevier.
Wilson, T. D., & Gilbert, D. T. (2008). Explaining away: A model of adaption. Perspectives on Psycho-
logical Science, 3(5), 370–386.

123